Dr. Mobeen Ur Rehman

Advancing Precision Agriculture (Funded By Khalifa University)

Our research focuses on revolutionizing precision agriculture by employing cutting-edge technologies. Through the integration of advanced computer vision and machine learning techniques, we address critical challenges in weed detection, disease identification, and yield prediction. By leveraging innovative methodologies beyond traditional approaches, our project aims to enhance overall crop management practices, contributing to sustainable and resilient agricultural ecosystems on a global scale.
 
 

Autonomous Detection of Aquatic Anomalies (Funded By Khalifa University)

Our research introduces an innovative autonomous system that utilizes computer vision, combining a convolutional model with remotely operated vehicles (ROVs) for accurate detection of anomalies in underwater imagery. Addressing challenges such as reflections and wave disturbances, the system demonstrates strong performance. Assessment in both controlled and real-world aquaculture environments highlights its capacity to improve safety measures, showcasing the effectiveness of computer vision in enhancing aquaculture management.
 
 
 

AI-Powered Epigenetic Analysis (Funded by NRF Korea)

Our project introduces cutting-edge artificial intelligence tools designed to identify epigenetic modifications, revolutionizing existing strategies for enhanced performance. We contribute by introducing new datasets and developing a single, versatile tool capable of identifying modified sites across the entire genome and in tissue-specific contexts. Our innovation lies in the generalizability of our in silico techniques, allowing for a comprehensive understanding of methylation's complex biological mechanisms. This platform establishes a direct linkage between methylation patterns and diseases, providing valuable insights through the tool's ability to learn intricate features.
 
 

Accurate Retinal Vessel Segmentation (In collaboration with ETRI Korea)

This research presents a state-of-the-art framework for precise retinal vessel segmentation. Leveraging a unique combination of advanced feature extraction, embedding, and dense multiscale feature fusion, our approach enhances the accuracy of identifying retinal vessels. These innovations hold significant promise for applications in medical imaging, particularly in the diagnosis and treatment of retinal diseases.
 
 
 

Advancements in Brain Tumor Segmentation from Multi-Modal MRI Images (Funded By JBNU, South Korea)

Brain tumors, often fatal, require precise segmentation for effective diagnosis. This research introduces an improved approaches for brain tumor segmentation in multi-modal MRI images, addressing existing limitations. Leveraging data pre-processing techniques, the study contributes to refining segmentation results and advancing the field of brain tumor analysis and diagnosis.
 
 
 

Blind image Quality Assessment (Funded By JBNU, South Korea)

Our project focuses on advancing blind image quality assessment, acknowledging the increased importance of screen content images alongside natural scenes in our technology-driven lives. The innovative approach considers both natural-scene and screen-content images without reliance on specific references or techniques. This research represents a substantial stride in blind image quality assessment, demonstrating promising results across diverse image types and scenarios.
 
 
 

AI-Driven Mitosis Detection in Breast Cancer

Our research focuses on revolutionizing mitosis detection in breast cancer, acknowledging the limitations of current methods reliant on manual assessments. We propose an innovative framework that leverages neural network concepts, reduced feature vectors, and diverse machine learning techniques for accurate cell classification. Emphasizing texture features crucial in mitosis detection, our framework showcases enhanced cancerous cell identification.